Criminal Justice - Is the Future Determined?

The accuracy numbers you get with these systems, it's technically called AUC for technologists and the audience, those numbers are around 70%. That's a bit better than random, which is 50%. But it's not nearly close to the kinds of things you might expect from a speech recognition system. These models are basically saying, the more time someone has been arrested in the past, the more likely they are to be arrested. And so that's where the problem comes in. Part of the problem is that these are highly imperfect predictors.